Fireside Chat: Supercharge Your Game art Production
summary
Layer AI is used to train custom game-art models from a studio’s own assets, claiming 3x productivity by turning weeks of outsourced background and prop work into minutes, with an initial test from ~20 uploaded images and 40–50 minutes to train a model that then generates images every 4–5 seconds. Production needs ~25 assets per model for a use case, upscales to high res, closes most outsourcing, scales across 60-person art teams, and flags isometric perspective limits plus copyright/ethics handled in private workspaces.
AI-written summary · about
chapters (19)
- 0:00Welcome
- 1:10Layer AI mission
- 2:38AI tools at Triple Dot
- 4:03Layer AI vs Mid Journey
- 5:29Triple Tile use case
- 5:50Changing art pipeline
- 6:35Copyright concerns
- 7:42Trained models benefits
- 9:08Speed and cost efficiency
- 10:14Isometric perspective challenges
- 11:19Adoption and ethics
- 12:03Scaling art teams
- 13:08Training models debate
- 15:47Fixing artifacts
- 16:29Artist familiarity
- 18:17Model training data
- 19:23Assets for training
- 20:06AI prompt guidelines
- 21:52Private workspaces
related talks
21mSession: AI Production for Game Studios – How it's Going And What To Expect In 2025.
34mAdvancing Game Development: AI-Driven Asset Creation and Concrete Workflow Integration
57mAImages - how neural networks will affect the production of games
20mStudio Speaks: The Reality of AI Adoption in Games
25mDesigning at the speed of AI
39mPanel: Hype Or A Real Force For Games – Where Are We With AI In 2025?
49mState of the Art: The Future of Visual Development.
35mPanel: All In Or Incremental? What's The Best Way To Integrate AI In Your Production?